Computer Laboratory for Granular Physics Studies
[ Home ] [ Software and Programming Resources ] [ Simulations ]
Granular materials involve a host of operations as part of the normal functioning of society. Example operations include the cultivation and shipping of huge volumes of construction materials (sand, gravel), agricultural and industrial processing of bulk goods (grains, nuts), and even the manipulation of soils upon which all buildings and infrastructure reside. Potentially vast improvements to society can be realized by gaining new insights into the behavior of granular materials, and corresponding gains in efficiency for said operations. The Computer Laboratory for Granular Physics Studies focuses on advancing the understanding of the fundamental nature of interactions between granular objects. Particular emphases are placed on characterizing the phenomenon of friction for granular soils and for modeling the introduction of structural objects into granular media across multiple scales. The laboratory aims to leverage such advances toward improving civil, industrial, and materials engineering processes.
The Computer Laboratory for Granular Physics Studies maintains access to two high-performance computing setups. The vehicle for carrying out all numerical simulations that require massively parallel processing (MPP) is a 384 normalized processing unit (NPU) compute cluster for sole use by the research team. The cluster administration is handled by the High Performance Computing (HPC) Center on the University of Florida campus. In this context, NPU denotes the use of a physical 64-bit architecture core with processing speeds ranging from 2.0 GHz to 3.3 GHz, drawn from a pool of 32,000 cores and a practically arbitrary number of compute nodes. For each node selected in carrying out a simulation, up to 2 GB of RAM can be utilized. For all numerical simulations that are better suited for shared memory processing (SMP), two 48-core Intel(R) Xeon® 2.30 GHz workstations, where each workstation is fitted with 64 GB of RAM.